On Friday 01 June 2012 19:24:45 Andreas Müller wrote: > On Fri, Jun 1, 2012 at 6:54 PM, Paul Eggleton > <[email protected]> wrote: > > I'd still like to rework this to use separate files. What if a line in the > > postinst script includes an = (which is not unlikely)? > > Yes after first usage (it is not possible on all my machines due to > pythongit version) and taking a look into the parsing scripts I now > understand your concern. It might cause non predictable results. Worst > case if there is a oe varaible name used in scripts. One question: > when outputting into separate files: isn't there a simple way to > attach the raw git diff output for these (maybe as option) in the > results of buildhistory-diff? Then we would have all information > together under one hood.
Definitely - if it doesn't already work once we've reworked it we can poke the code in lib/oe/buildhistory_analysis.py to just report the diff of the blobs (we already do this for other classes of files, some of the image ones IIRC). Cheers, Paul -- Paul Eggleton Intel Open Source Technology Centre _______________________________________________ Openembedded-core mailing list [email protected] http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
